Dummy variable

Up till now we have dealt exclusively with the

variables which can measures in quantitative
terms. But sometimes variables which we
consider important are of qualitative
character. The presence of such variables
cannot be measured quantitatively, but can
only be noted whether the given character is
present or not. For example, suppose that we
want to explain the consumption behavior of
different households. In addition to the level
of disposable income,

Dummy variable
we may believe that consumption
depends on order of other character
i.e. the presence or absence of
children, posses or not own house,
the literacy of the head of household,
religion etc. Since we cannot
measure these character but we can
assign 1 to presence and zero to
absence. In this way the attribute is
transferred to a variable known as

 Dummy variable are commonly used

in econometric research for
qualitative factor, Such as profession,
religion, sex, region etc. Dummy
variable can be used in regression
models just as easily as quantitative
variables.

 In order to explain consider the

example of simple linear regression
model in which the explanatory
variable is represented by dummy
variable. Assume that in a factor
salary offered to a worker depends
on literacy, i.e.

If
If

is salary offered
1 if the worker is educated
= 0 if the worker is uneducated
then
then

 The intercept term

gives mean salary of
the worker having low qualification and
the slope coefficient
tells us by how
much the mean salary of worker having
qualitative different from the mean salary
of low qualification on workers.
A test of hypothesis
is equivalent to
the test that there is no difference
between the mean salaries of the workers.

 The model which contain explanatory

variables that are exclusively dummy
variable are called analysis of
variance (ANOVA) model. In most of
economic research the regression
model contains a admixture of
qualitative and quantitative variables
are called ANCOVA model.

 Where Ci
H.H. yi
D1
D2

=0
=1
=0
=1

is the consumption of head

is the income of head of H.H.
if H.H has no childern.
if H.H has childern.
if H.H has no house.
if H.H has house.

D3

=0 if the head H.H is

illiterate.
=1 if the head H.H is literate
If a person has no children, has no
own house and illiterate then the
mean consumption function

 Uses of Dummy variables

Use of dummy variable for measuring the shift of a
function over time
A shift of a function implies that the constant intercept
changes in different period, while other coefficient
remains constant. Such type of shift can be examined
by introducing dummy variables in function under study.
For example we wish to study the aggregate
consumption function for the period 1910-1960. There
were two wars, partition of the countries and also
depression, hence the condition were not normal, during
war times for given income we might expect downward
shift of consumption.

 Use of Dummy variable for

measuring change in
(slope)parameter over time
It is known that over long period of
time or abnormal year , not only the
function is shifted but also their slope
will be expected to change. The
change in parameters may be
captured by introducing appropriate
dummy in the model.

 Features of Dummy variables in Regression model

One dummy variable is sufficient to distinguished
two categories. We introduce a dummy variable with
zero for one category and one for other category.
The category which is assigned the value zero is
often referred to as the base or control category.
If we have more than two categories, let say K then
(k-1) dummy variables will be used. The value one
of each dummy variable will be used for each
category other than base category and the zero of
all dummy will represent the base category.

 Intercept Dummy

 Slope Dummy
Example

 Question 1: Using your EAEF data

set, regress S on ASVABC, SM, SF and
MALE, a dummy variable that is 1 for
male respondents and 0 for female
ones. Interpret the coefficients and
perform t tests. Is there any evidence
that the educational attainment of
males is different from that of
females?

 Question 2: Does ethnicity affect educational

attainment? In your EAEF data set you will find
the following ethnic dummy variables:
ETHHISP 1 if hispanic, 0 otherwise
ETHBLACK 1 if black, 0 otherwise
ETHWHITE 1 if not hispanic or black, 0 otherwise.
Regress S on ASVABC, MALE, SM, SF, ETHBLACK
and ETHHISP. (In this specification ETHWHITE has
been chosen as the reference category, and so it is
omitted.) Interpret the regression results and
perform t tests on the coefficients

 Question 3. Are earnings subject to

ethnic discrimination? Using your
EAEF data set, regress LGEARN on S,
ASVABC, MALE, ETHHISP and
ETHBLACK. Interpret the regression
results and perform t tests on the
coefficients.

 Question No 4: Is the effect of the ASVABC

score on educational attainment different for
males and females? Using your EAEF data set,
define a slope dummy variable MALEASVC as
the product of MALE and ASVABC:
MALEASVC = MALE*ASVABC
Regress S on ASVABC, SM, SF, ETHBLACK,
ETHHISP, MALE and MALEASVC, interpret the
equation and perform appropriate statistical
tests